你好！
当我使用UniversalDependencyConverter模块从解析树生成UD时，遇到了意外的输出。正如你所看到的，第18行在下面的命令输出中没有头部(它有一个_代替)。你知道这可能是什么原因吗？

$ CORENLP_HOME="/home/gaguil20/tools/stanford-corenlp-full-2018-10-05"
$ java -cp "$CORENLP_HOME/*" \
       -mx5g edu.stanford.nlp.trees.ud.UniversalDependenciesConverter \
       -outputRepresentation enhanced++ \
       -treeFile parsing.txt

1	What	what	PRON	WP	_	3	nsubj	3:nsubj	_
2	's	be	VERB	VBZ	_	3	cop	3:cop	_
3	interesting	interesting	ADJ	JJ	_	14	dep	14:dep	_
4	,	,	PUNCT	,	_	14	punct	14:punct	_
5	I	I	PRON	PRP	_	6	nsubj	6:nsubj	_
6	mean	mean	VERB	VBP	_	18	parataxis	14:dep|18:parataxis	_
7	,	,	PUNCT	,	_	14	punct	14:punct	_
8	reading	read	VERB	VBG	_	14	dep	14:dep	_
9	the	the	DET	DT	_	10	det	10:det	_
10	article	article	NOUN	NN	_	8	dobj	8:dobj	_
11	,	,	PUNCT	,	_	14	punct	14:punct	_
12	basically	basically	ADV	RB	_	14	dep	14:dep	_
13	a	a	DET	DT	_	17	det:qmod	17:det:qmod	_
14	lot	lot	NOUN	NN	_	13	mwe	13:mwe	_
15	of	of	ADP	IN	_	13	mwe	13:mwe	_
16	the	the	DET	DT	_	17	det	17:det	_
17	celebrities	celebrity	NOUN	NNS	_	0	root	0:root	_
18	paid	pay	VERB	VBN	_	_	_	_	_
19	by	by	ADP	IN	_	20	case	20:case	_
20	him	he	PRON	PRP	_	18	nmod:by	18:nmod:by	_
21	to	to	PART	TO	_	22	mark	22:mark	_
22	attend	attend	VERB	VB	_	18	xcomp	18:xcomp	_
23	charity	charity	NOUN	NN	_	24	compound	24:compound	_
24	functions	function	NOUN	NNS	_	22	dobj	22:dobj	_
25	.	.	PUNCT	.	_	14	punct	14:punct	_

parsing.txt文件包含以下解析树(取自OntoNotes 5.0的Broadcast News部分):

(TOP
    (FRAG
        (SBAR
            (WHNP
                (WP What))
            (S
                (VP
                    (VBZ 's)
                    (ADJP 
                        (JJ interesting)))))
        (, ,)
        (PRN
            (S
                (NP
                    (PRP I))
                (VP
                    (VBP mean))))
        (, ,)
        (S
            (VP
                (VBG reading)
                (NP
                    (DT the)
                    (NN article))))
        (, ,)
        (ADVP
            (RB basically))
        (NP
            (NP
                (DT a)
                (NN lot))
            (PP
                (IN of)
                (NP
                    (DT the)
                    (NNS celebrities))))
        (VP
            (VBN paid)
            (PP
                (IN by)
                (NP
                    (PRP him)))
            (S
                (VP
                    (TO to)
                (VP
                    (VB attend)
                    (NP
                        (NN charity)
                        (NNS functions))))))
    (. .)))

非常感谢你的帮助，并感谢你提供这样一个伟大的工具！
祝好，
古斯塔沃

[更新]:我在Red Hat 7.6上运行此命令，OpenJDK版本为1.8